The effect of SEM imaging on the Ar/Ar system in feldspars

Complex microtextures form in K-feldspar crystals as they cool and are affected by deuteric alteration. This complex structure is the cause of variable closure temperatures for Ar-Ar, a phenomenon which has been utilized in multi domain diffusion (MDD) modelling to recover thermal histories [1]. However, there has been substantial controversy regarding the precise interaction between feldspar microtextures and Ar-diffusion [2,3]. A number of studies have addressed this issue using coupled SEM imaging and Ar/Ar UV laser ablation microprobe (UV-LAMP) analysis on the same sample, to enable direct comparison of microtextures with Ar/Ar age data [4]. Here we have tested the idea that SEM work may affect Ar/Ar ages, leading to inaccurate results in subsequent Ar/Ar analyses. Three splits of alkali feldspar from the Dartmoor Granite in SW England were selected for Ar/Ar UV-LAMP analysis. Split 1 (“control”) was prepared as a polished thick section for Ar/Ar analysis. Split 2 (“SEM”) was prepared as a polished thick section, was chemically-mechanically polished with colloidal silica and underwent SEM imaging (uncoated) and focussed ion beam (FIB) milling (gold coated); electron beam damage in the SEM was maximised by leaving the sample at high magnification for eight minutes. Split 3 (“Etch”) is a cleavage fragment that was etched with HF vapour and underwent low to moderate magnification SEM imaging. The control split gave a range of laser-spot ages consistent with the expected cooling age of the granite and high yields of radiogenic 40Ar* (>90%). The area of the “SEM” split that experienced significant electron beam damage gave younger than expected ages and 40Ar* yields as low as 57%. These are interpreted as a combination of implantation of atmospheric Ar and local redistribution of K within the sample. The area of “SEM” that underwent FIB milling gave ages and 40Ar* yields comparable to the control split, suggesting that the Au-coat minimises FIB damage and that colloidal-polishing and low-magnification SEM imaging do not affect the Ar/Ar system. The “Etch” split gave younger than expected ages and 40Ar* yields as low as 58%, suggesting that HF etching also disrupts the Ar/Ar system. These results suggest that SEM techniques involving intense electron bombardment of an uncoated sample, such as charge contrast imaging and electron backscatter diffraction (EBSD), may disrupt the Ar/Ar system in the sample, leading to spurious results. Etching samples with HF, as is often done for routine Ar/Ar preparation of volcanic phenocrysts, introduces atmospheric Ar and may result in differential loss or gain of K and Ar isotopes, leading to spurious results. References [1] Lovera and Richter, 1989, J. Geophys. Res. 94, 17917-17935. [2] Parsons. et al., 1999, Cont. Min. Pet. 136, 92-110. [3] Mark et al., 2008, Geochim. Cosmochim. Acta, 72 2695-2710. [4] Reddy et al., 2001, Cont. Min. Pet., 141 186-200.